Re: [ANN] tnetstrings.lua 1.0.0

[Josh Simmons <simmons.44@...>]

On Sun, Jun 5, 2011 at 11:27 PM, Florian Weimer <fw@deneb.enyo.de> wrote:
> * Josh Simmons:
>
>> * Trivial to parse in every language without making errors.
>
> Apparently, this is not entirely true:
>
>            val, ext = parse(ext)
>
>            -- If val is nil, then ext is actually an error message.
>            if not val then
>                return val, ext
>            end
>
>            local val, ext = parse(ext)
>            if not val then
>                return nil, ext
>            end
>

Haha, fixed.

> Your implementation does not round-trip, either.

It seems to, according to the tests.

>
>> * Resistant to buffer overflows and other problems.
>
> Your implementation results in near-quadratic run time in terms of the
> input string length.
As far as I can tell, there's not a huge amount that can be done about
that. But I'm open to suggestions.

>
>> * Fast and low resource intensive.
>
> The format requires buffering all data before decoding can start.
> This means that decoding arbitrary messages requires unbounded
> storage.
You can't really get around that while still maintaining "Makes no
assumptions about string contents" and at any rate it's more a goal to
be able to predict that than actually handle it.

>
>> * Transport agnostic, so it works with streams, messages, files,
>> anything that's 8-bit clean.
>
> The format is non-streaming, so it cannot itself serve as a transport
> for byte streams.

It has to be chunked indeed, but very rarely is a true streaming
system actually implemented. Especially not in the context the
protocol was designed for, internal communication for a web server
(mongrel2).